PROJECT SUMMARY More than 5.7 million American adults experience heart failure (HF), and prevalence is expected to increase by 46% in the next two decades. Total costs of HF care exceed $30 billion annually. Patients with HF are at high risk for poor outcomes, including hospitalization, disability, and death. A novel intervention for lowering costs and improving quality of care for Medicare beneficiaries is the implementation of Accountable Care Organizations (ACOs). ACOs consist of groups of doctors and other health care entities that come together voluntarily to provide their patients with coordinated, high-quality care. If this results in lower costs, the ACO shares the savings with the Medicare program. ACOs are simultaneously required to meet quality metrics to ensure that cost reductions are the result of improvements in population health (and not the result of denying services to beneficiaries). As of 2017, the Medicare Shared Savings Program (MSSP), a large Medicare program that provides incentives for ACO formation, included more than 9 million Medicare patients across more than 400 ACOs nationwide. Preliminary data suggests there is variation across ACOs in outcomes for patients with HF. We hypothesize that this variation is related to differences in the organization and delivery of care within ACOs. For example, use or implementation of care management may vary across ACOs, leading to differences in outcomes. Our long-term goal is to identify effective delivery-system interventions that improve the health and outcomes of patients with HF. The objective of this proposal is to identify the strategies and contextual factors that lead some ACOs to achieve better risk-standardized outcomes including: lower acute admission rates (RSAARs), fewer days in skilled nursing care (RSSNFs), and lower mortality rates (RSMRs). Aim 1 will examine outcomes across ACOs and identify patterns in the trajectories of outcomes over time. Each ACO will be assigned a ?phenotype? based on trends in risk-standardized outcomes. We will then examine the association between phenotypes with ACO characteristics. In Aim 2, we will conduct qualitative interviews with a sample of ACOs taken from high- and low- performing phenotypes with the goal of refining our understanding of the factors and strategies that allow some ACOs to achieve better outcomes. For Aim 3, we will conduct a survey of all remaining MSSP ACOs (those not sampled in Aim 2) to examine associations between identified strategies and outcomes using multinomial regression. The results of this work will inform health system leaders and policymakers as they implement care management strategies within ACOs and consider changes to the MSSP more broadly. These activities are highly responsive to the National Heart Lung and Blood Institute?s Strategic Goals and Objectives, specifically objective 6, which aims to ?optimize translational, clinical, and implementation research to improve health and reduce disease.?